Operand forwarding (or data forwarding) is an optimization in pipelined CPUs to limit performance deficits which occur due to pipeline stalls.[1][2] A data hazard can lead to a pipeline stall when the current operation has to wait for the results of an earlier operation which has not yet finished.

Example

edit
ADD A B C  #A=B+C
SUB D C A  #D=C-A

If these two assembly pseudocode instructions run in a pipeline, after fetching and decoding the second instruction, the pipeline stalls, waiting until the result of the addition is written and read.

Without operand forwarding
1 2 3 4 5 6 7 8
Fetch ADD Decode ADD Read Operands ADD Execute ADD Write result
Fetch SUB Decode SUB stall stall Read Operands SUB Execute SUB Write result
With operand forwarding
1 2 3 4 5 6 7
Fetch ADD Decode ADD Read Operands ADD Execute ADD Write result
Fetch SUB Decode SUB stall Read Operands SUB: use result from previous operation Execute SUB Write result

In some cases all stalls from such read-after-write data hazards can be completely eliminated by operand forwarding:[3][4][5]

With operand forwarding (enhanced)
1 2 3 4 5 6
Fetch ADD Decode ADD Read Operands ADD Execute ADD Write result
Fetch SUB Decode SUB Read Operands SUB: use result from previous operation Execute SUB Write result

Technical realization

edit

The CPU control unit must implement logic to detect dependencies where operand forwarding makes sense. A multiplexer can then be used to select the proper register or flip-flop to read the operand from.

See also

edit

References

edit
  1. ^ "CMSC 411 Lecture 19, Pipelining Data Forwarding". University of Maryland Baltimore County Computer Science and Electrical Engineering Department. Retrieved 2020-01-22.
  2. ^ "High performance computing, Notes of class 11". hpc.serc.iisc.ernet.in. September 2000. Archived from the original on 2013-12-27. Retrieved 2014-02-08.
  3. ^ Gurpur M. Prabhu. "Computer Architecture Tutorial". Sections "Forwarding". and "Data Hazard Classification".
  4. ^ Dr. Orion Lawlor. "Pipelining, Pipeline Stalls, and Operand Forwarding".
  5. ^ Larry Snyder. "Pipeline Review".
edit